Accuracy
You are probably wearing it wrong
Before you decide your band is inaccurate, check where it is sitting. Position and strap tension change the quality of the signal more than the difference between most devices on the market.
What the sensor needs
Every fix in this article is a way of delivering one of those three. None of them costs anything.
Before you conclude your band is inaccurate, check where it is sitting. A few centimetres of position and a notch of strap tension change the quality of the signal more than the difference between most devices on the market.
This is the least glamorous article we will publish and probably the most immediately useful. None of it costs anything.
How the sensor actually works
The green light on the underside is doing photoplethysmography. The device shines light into your skin and measures how much comes back. Blood absorbs that wavelength, so with each heartbeat — as blood volume in the capillaries under the sensor rises and falls — the amount returning changes slightly. The device counts those fluctuations.
Which tells you what the sensor needs: consistent, close contact with still skin, and no outside light. Everything below follows from those three requirements.
Position: above the wrist bone
The most common mistake is wearing the band like a watch — sitting low, resting on the wrist bone.
What goes wrong
- Sitting on the wrist bone, so it rocks
- Sliding down toward the hand in use
- Loose enough to admit daylight
- Tight enough to leave a mark
- Sensor over a tattoo
What to do
- One to two fingers above the wrist bone
- Snug, so it does not travel
- Contact all the way round
- Moves with the skin, not against it
- Try the other wrist if readings are odd
About a finger or two above the bone puts the sensor on flatter tissue with better blood flow and less movement, and that is where the signal is cleanest. Most manufacturers say the same thing, and almost nobody does it.
For exercise, move it further up and tighten a notch. Then loosen it again afterwards. During activity the arm accelerates hard, and any gap between sensor and skin becomes motion artefact. This single adjustment fixes more “my heart rate readings are nonsense during workouts” complaints than anything else.
Tension: snug, not tight
There is a narrow band of correct tightness, and both ways out of it cause problems.
The sensor lifts off the skin as you move, admitting ambient light and losing the signal. Produces dropouts, implausible spikes, and heart rate that lags behind effort.
You can feel it is there but it is not uncomfortable. It does not slide when you turn your wrist over, and it leaves no lasting mark when removed.
Compressing the tissue reduces the blood flow the sensor is trying to read. Strap tightness causing skin compression is a known source of measurement error — and it is the main cause of strap rash.
The test: turn your wrist over and give the band a gentle push. If it travels, tighten one notch. If removing it leaves an indentation that takes minutes to fade, loosen one.
What good fit cannot fix
Fit matters, but it is worth being clear about what remains after you have got it right.
- Rapid intensity changes. This is the hard case for optical sensing, and it is well documented — error increases during dynamic changes in activity intensity. Intervals and sprints are where wrist readings diverge most from a chest strap, and no amount of adjustment closes that gap entirely.
- Cold. Peripheral blood flow drops in the cold, which weakens the signal at the wrist. Winter outdoor sessions are genuinely harder for the sensor.
- Tattoos. Dense ink absorbs the light the sensor depends on. If the sensor sits over heavy tattooing, the other wrist is the fix.
- Skin tone. Worth stating carefully, because the evidence is mixed. A systematic review found four studies showing no skin-tone effect, four showing reduced accuracy for darker skin, and two mixed. A 2025 study reported that skin tone did not affect error, while device and activity condition did. The fair summary is that activity intensity matters more than skin tone for most devices, and that the research has not settled.
Movement is the enemy of an optical sensor. Almost every fix is a way of reducing it.
Wearing it overnight
Overnight is where a band earns its keep — resting heart rate, HRV and readiness all come from those hours. It is also where people get the fit wrong in the opposite direction.
-
Slightly looser than daytime, not loose
Your wrist swells a little overnight. One notch looser than your training fit is usually right — still in contact, no longer compressing.
-
Keep it above the bone
The same position rule applies. A band that has slid onto the wrist bone by 3am is reading the worst available spot during the most valuable hours.
-
Same wrist, every night
Baselines are comparisons. Swapping wrists introduces a difference that looks like a change in you.
-
Keep the sensor clean
Sweat, soap residue and moisturiser leave a film that scatters light. A wipe with a dry cloth every few days is enough, and it matters more than people expect.
-
Let the skin breathe
Take it off while charging and give your wrist that time dry and uncovered. In a humid climate this is what prevents the irritation that makes people stop wearing it.
A two-minute check
If you suspect your data is wrong, do this before blaming the device.
| What you see | Usually | Try |
|---|---|---|
| Heart rate lags behind effort | Loose during movement | Move up the arm, tighten a notch |
| Wild spikes in intervals | Motion artefact | Tighten; use a chest strap for sprints |
| Gaps in overnight data | Slid onto the wrist bone | Reposition above the bone |
| Strap rash or indentation | Too tight | Loosen a notch; dry the wrist daily |
| Weak readings outdoors in winter | Cold peripheral circulation | Cover the band with a sleeve |
| Consistently odd readings | Tattoo or sensor film | Clean the sensor; try the other wrist |
Most people who think their band is broken have a strap problem.
Why fit was a design question for us
Removing the screen let Mesh 2.0 be lighter and slimmer, and weight is the main reason bands slide down the arm and get taken off at night. A band you forget you are wearing is a band that stays in the right position — which is most of what accuracy actually depends on.
That is the thread connecting taking the screen off, the battery life, and the data quality: they are the same decision viewed from different angles. Read the full Mesh 2.0 specification.
Questions people actually ask
Where should a fitness band sit on your wrist?
About one to two fingers above the wrist bone, not resting on it. That puts the sensor on flatter tissue with better blood flow and less movement, which is where the optical signal is cleanest.
For exercise, move it a little further up the arm and tighten one notch, then loosen it again afterwards.
How tight should a fitness tracker be?
Snug enough that it does not travel when you turn your wrist over, loose enough that removing it leaves no lasting indentation.
Too loose lets the sensor lift off the skin and admit ambient light, causing dropouts and spikes. Too tight compresses the tissue and reduces the blood flow the sensor is trying to read — strap tightness causing skin compression is a documented source of measurement error.
Why is my heart rate wrong during workouts?
Usually fit, and partly physics. A band that is loose or sitting low becomes a motion-artefact generator once your arm starts accelerating. Moving it further up the arm and tightening a notch fixes most cases.
What fit cannot fix is rapid intensity change: optical sensors are documented to lose accuracy during dynamic changes in activity intensity. For sprints and intervals, a chest strap is a genuinely different class of measurement.
Should I wear my fitness band to bed?
Yes, if you want resting heart rate, HRV or a readiness score — all of them are built from overnight measurement, and nights you do not wear it degrade the baseline the other nights are compared against.
Wear it one notch looser than your training fit, since wrists swell slightly overnight, keep it above the wrist bone, and use the same wrist every night so the comparison stays valid.
Do tattoos affect heart rate tracking?
They can. Dense ink absorbs the light the optical sensor depends on, so a sensor sitting over heavy tattooing may read poorly or not at all.
The fix is simply to wear the band on the other wrist, or in a position where the sensor is on untattooed skin.
Does skin tone affect accuracy?
The evidence is mixed and worth reporting honestly. A systematic review found four studies showing no skin-tone effect on accuracy, four showing reduced accuracy for darker skin, and two with mixed results. A 2025 study reported that skin tone did not affect measurement error while device and activity condition did.
The fair summary is that activity intensity and rapid heart rate changes matter more than skin tone for most devices, and that the research has not settled the question.
Sources
- Investigating sources of inaccuracy in wearable optical heart rate sensors, npj Digital Medicine, 2020 — measurement error increasing during dynamic changes in activity intensity.
- Reliability and accuracy of the Fitbit Charge 4 photoplethysmography heart rate sensor in ecological conditions: validation study, 2024 — identifies device placement on skin, strap tightening causing skin compression, skin tone, and activity type and intensity as factors affecting heart rate recording.
- Accuracy of heart rate measurement with wrist-worn wearable devices in various skin tones: a systematic review — four studies showing no skin-tone effect, four showing decreased accuracy for darker skin, two with mixed results.
- Validity of heart rate measurements in wrist-based monitors across skin tones during exercise, PLOS ONE, 2025 — reports a statistically significant interaction between skin tone and exercise intensity; separate 2025 work reports skin tone not affecting error while device and activity condition did.
Findings are quoted as reported by the cited studies, which cover specific device models and protocols. The research on skin tone and photoplethysmography accuracy is genuinely inconsistent and is reported here as such rather than resolved. Guidance on position and strap tension reflects manufacturer advice and the factors identified in validation work; it is general information about getting better readings, not medical advice.
Mesh 2.0
Wear less.Know more.
A screenless band that tracks your heart rate, sleep and BioFuel — then stays out of the way. Available now across Pakistan.